The Expression, Purification, and Characterization of pGEM-gbr22 Recombinant Protein
Introduction:
Many different types of organisms have been used to synthesize proteins from another organism of interest. The relatively harmless strain of competent bacterial E. Coli BL21 (DE3) has consistently proven itself as one of the optimal hosts for expression of proteins of higher level organisms due to a rapid rate of bacterial replication and a lack of Ion and ompT proteases and its compatibility with the T7 lacO promoter system [1]. Recombinant protein plasmids have often been tagged with a hexa-histidine affinity tag for future ease of purification of the protein, and a ampicillin resistant gene has often been inserted into the plasmid as well [2]. The recombinant plasmid may then be inserted into the E. Coli BL21 (DE3) bacterial genome, and the bacteria can be cultivated to express the protein from the plasmid. After protein expression, the bacterial cells could then become lysed, and the protein of interest could run through a column containing Ni-NTA resin so that the hexa-histidine affinity tag would bind to the resin and the protein of interest would then be successfully isolated and purified [3]. For a basic introduction into protein expression and purification, a visibly purple and modified (ampicillin resistant gene and insertion of hexa-histidine tag) pGEM-gbr22 Great Barrier Reef coral protein was expressed in the E. Coli BL21 (DE3) bacteria. After cultivation and successful protein production, the pGEM-gbr22 protein of interest was purified through the use of a Ni-NTA resin column, and then the protein was later characterized through the use of gel electrophoresis. It can be hypothesized that the the pGEM-bgr22 protein will ultimately be isolated from its host E. Coli bacteria, but it can never be fully isolated from some of the contents of the host bacteria in which it grew in.
Materials & Methods:
While in proximity of a lit gas Bunsen burner, 25 microliters of competent E. Coli BL21 (DE3) bacterial cells were added into one transformation tube, and the same amount of bacterial cells were added into another transformation tube. 1.29 microliters of pGEM-gbr22 plasmid was then carefully pipetted into only one of the transformation tube. After the two transformation tubes were placed on ice for 30 minutes, the two tubes were heat shocked for exactly 45 seconds in a 42 degree Celsius water bath. The two tubes were then placed in ice for another 2 minutes. 200 microliters of SOC media was then pipetted into the two tubes, then the tubes were placed in a water bath incubator (~250 RPM) for 30 minutes at 37 degrees Celsius. Next, 5 colirollers were placed onto a agar ampicillin plate labeled 'DNA', and 5 colirollers were placed onto a seperate agar ampicillin plate labeled 'No DNA (Control)'. After the transformation tubes finished their 30 minutes in the water bath incubator, the 50 microliters of the bacteria/SOC mixture was pipetted from the tube with the plasmid into the 'DNA' plate. 50 microliters of the bacteria/SOC mixture was then pipetted from the tube without the plasmid into the 'No DNA (Control)' plate. The mixtures were then rolled around in their respective plates, the colirollers were then carefully poured out, and then the plates were placed in a 37 degree Celsius incubator overnight. A fun plate with microorganisms from the men's restroom was also created and incubated overnight. On the next morning, the plates were taken from the incubator, and 0.01 ml of ampicillin stock (50 mg/ml) was added to two 14 ml round bottom conical tubes with 5 ml of LB in it. Two sterile small pipette small were then used to gently swipe and pick up each a single colony of transformed bacteria from the 'DNA' plate. One tip (with the bacterial colony on it) was then entirely submerged in one of the tubes of LB and ampicillin media, and the other tip was then entirely submerged in the second tube of LB and ampicillin media as a backup. These two conical tubes were then placed in a shaking water bath incubator (~250-300 RPM) at 37 degrees Celsius for 8 hours. After 8 hours, the two tubes were removed from the water bath, and one 125 ml Erlenmeyer Flask was filled with 25 ml of LB media and of 0.05 ml ampicillin. Exactly 0.625 ml of the culture from one of the tubes (the most visibly purple tube) that has just been removed from the water bath was then pipetted into the Erlenmeyer Flask. Foil was placed over the top of the flask, and the flask was placed back into the shaking water bath incubator at the same settings as before, but 16-24 hours of growth was then allowed. After the allotted 16-24 hours of incubation, the now extremely visibly purple flask was removed from the incubator, and 500 ml of the mixture in the flask was pipetted into a 1.7 ml Eppendorf tube and saved in 4 degree Celsius refrigerator labeled as Sample 1. The bacteria and the media from the flask were then poured in its entirety into a 50 ml conical tube. After securely capping the tube, it was then placed into the Allegra X-15 (Beckman Coulter, Inc., Brea, CA) benchtop centrifuge with another equally weighted conical tube of the same mass in opposite position for balancing purposes. The tube was centrifuged at 5,000 RPM for 10 minutes at 4 degrees Celsius. After centrifuging, the liquid supernatant was removed from the tube and was decanted into a bleach waste container. The remaining purple cell pellet was pipetted up and down with 2.5 ml of 1x PBS solution to form a homogenous suspension. 50 microliters of lysozyme was then added to the conical tube was well, and once vortexed, it was placed in a -20 degree Celcius freezer. After a week or so, the 50 ml conical tube containing the lysozyme and the 2.5 ml of suspended bacterial cells was thawed, and then 2 microliters of Benzonase (Sigma-Aldrich, St. Louis, MO) was added to the conical tube. It was mixed by inversion for about 15 minutes at room temperature. After this, a 1 ml of the mixture was pipetted in the conical tube into two 1.7 ml Eppendorf tubes, and the pipette the rest into a third 1.7 ml Eppendorf tube. The three Eppendorf tubes were then centrifuged in the small centrifuge for 20 minutes at 14,000 RPM at 4 degrees Celsius. After centrifuging, 50 microliters of the supernatant from one of the tubes was stored and labeled as Sample 2. The bacterial cell debris was the resulting pellet from centrifuging, so the liquid supernatant (containing soluable proteins) from each of the three tubes was extracted through pipetting, and then it was placed in a clean 15 ml conical tube. A 10 ml Wash buffer containing 1.0 ml of 1x PBS and 0.2 ml of imidazole, and a 10 ml Elution buffer containing 1.0 ml of 1x PBS and 2.5 ml of imidazole were then created and kept on ice. The supernatant was then filtered through a PES syringe filter (Membrane Solutions, Plano, TX). A purification column was then set up by first adding 0.5 ml of Ni-NTA resin/buffer mixture into the conical tube of supernatant. Through occasional inversion mixing, this new mixture was incubated at room temperature for 15 minutes. A 20 ml purification chromatography Econo column (Bio-Rad, Hercules, CA) was then set up on a ring stand with a ice bucket below it. The column was first rinsed with nanopure water, then the supernatant with the Ni-NTA resin/buffer was ran through the column, allowed to settle for about 5 minutes, and then it was dripped into a 10 ml round bottom conical tube labeled 'waste'. A 50 microliter sample of waste solution was taken and labeled Sample 3 and stored in a 4 degree Celsius refrigerator. It was crucial to leave a small amount of the 'waste' solution in the column so that not all of it flowed out of the column (the protein of interest settled to the bottom of the column). After re-plugging the column, 5 ml of the Wash solution was ran through the column into a new 10 ml conical tube labeled 'Wash' after about 5 minutes of letting the Wash solution settle. A 50 microliter sample of Wash solution was taken and labeled as Sample 4 and stored in a 4 degree Celsius refrigerator. After leaving a small amount of solution in the column, 5 ml of the Elution solution was ran through the column in the same way as before into a 'Elution 1' 15 ml conical tube. A 50 microliter Sample 5 was taken from this Elution 1 tube and stored in a 4 degree Celsius refrigerator. Another 5 ml of Elution solution was ran through the column into a 'Elution 2' 15 ml conical tube. A 50 microliter Sample 6 was taken from this Elution 2 tube and stored in a 4 degree Celsius refrigerator. To strip the Ni-NTA from the column, 10 column volumes (cv) of water, then 10 cv's of 0.5 NaOH, then 10 cv's of water were ran through the column one at a time, and then the column was stored in a 4 degree Celsius refrigerator with 1 ml of 30% ethanol in nanopure water sitting on the bottom of the column. In order to estimate the concentration of the pGEM-gbr22 purple protein, a Nanodrop (Thermo Scientific, Wilmington, DE) spectrophotometer was used to measure the absorbance of the protein at 280 nm and the maximal wavelength of the protein (574 nm). This was done by first blanking the spectrophotomer with 2 microliters of nanopure water, then blanking the instrument with 2 microliters of Elution solution buffer, and then measuring the protein's absorbance by adding 2 microliters of Elution 1 to the spectrophotometer (ensuring a Kimwipe was used to scrub the pedestal after each step). In the process of characterizing the protein, 10 microliters of a 6x gel loading buffer was added to the saved Samples 2-6 and pipetted up and down to mix well. Sample 1 was centrifuged at 5,000 RPM for 5 minutes using a small centrifuge with a 500 microliter Eppendorf tube of water as a counterbalance. The supernatant was removed after centrifuging, and then 200 microliters of water and 40 microliters of the 6x gel loading buffer were added to the cell pellet and mixed to create a homogeneous solution. Then all the samples (1-6) were placed on a 95 degree Celsius heat block for 5 minutes and then centrifuged at 5,000 RPM for 2 minutes. An SDS electrophoresis module was then assembled and a 1x TGS buffer was used to fill the tank. After the 10 wells were cleaned using a small needle and about 2 ml of the 1x TGS buffer and the tape on the bottom of the gel was removed, 7 microliters of a Fermentas prestained protein ladder (Thermo Fisher Scientific, Waltham, MA) was pipetted into the first well, and then wells 2-7 were filled with 20 microliters of pipetted Samples 1-6. Wells 8-10 were filled with a partner's Samples 4-6. The electrophoresis module was then ran for 25 minutes at 200 Volts. When finished, the gel was removed from its casing and allowed to float in a small plastic dish of nanopure water. The SDS residue on the gel was then removed by washing the gel three times in a orbital shaker with about 100 ml of nanopure water for five minutes each time. After the water was removed from the plastic container, an Imperial protein stain (Pierce Biotechnology, Rockford, IL) was then added to completely cover the gel, and then the container was placed back on the orbital shaker for about an hour. After an hour, the stain was poured back into the container from which it originated, and the gel was washed twice (as described previously), and a folded kimwipe was added to the container filled with clean water and gel for overnight orbital shaking. On the next day, the gel was removed from the plastic container, placed ontop a rectangle of Whatman filter paper (GE Healthcare, Maidstone, UK), and a piece of rectangular cellophane was then placed on top of the gel. The gel was then placed on a drying bed for 1.5 hours at 75 degrees Celsius on the Gradient cycle.
Results: Figure 1: Image of "Fun Plate" bacterial, virus, and fungal colonies. Microorganisms collected from men's restroom toilet handle. Incubated in 37 degree Celsius incubator for about 24 hours. Ampicillin not present on agar of plate.
Figure 2: Image of E. Coli BL21 (DE3) bacterial colonies incubated for about 24 hours in 37 degree Celsius incubator in the presence of ampicillin added to the agar. Very few colonies developed due to the ampicillin's antibiotic effect that killed off most of the bacterial colonies during incubation.
Figure 3: Image of E. Coli BL21 (DE3) bacterial colonies incubated for about 24 hours in 37 degree Celsius incubator. These bacterium have been manipulated to express an ampicillin resistant purple Great Barrier Reef coral gene (pGEM-gbr22) in the form of a protein. There are numerous bacterial colonies despite there being ampicillin in the agar because the bacteria with the included coral gene now express an ampicillin resistant gene. Most other bacterial colonies that may have developed in the plate died during incubation due to the presence of ampicillin in the agar.
Figure 4: Image of E. Coli BL21 (DE3) bacteria with expressed pGEM-gbr22 protein resting in an Erlenmeyer Flask mixed with 25 ml LB broth and ampicillin media after about 24 hours of incubation in 37 degree Celsius shaking water bath incubator.
Figure 5: Image of spun down (from Allegra X-15 benchtop centrifuge) E. Coli BL21 (DE3) bacteria with expressed pGEM-gbr22 protein pellet with the supernatant removed from the 50 ml conical tube. Wet cell pellet weighed about 0.24 g.
Figure 6: Image of Elution 1 (bulk of purified pGEM-gbr22 protein washed down by Elution buffer solution in Ni-NTA column) and Elution 2 (remnants of purified pGEM-gbr22 protein that did not drip into Elution 1 tube from the Ni-NTA column) of purified pGEM-gbr22 protein in two 15 mL conical tubes.
Figure 7: Image of Nanodrop spectrophotometer absorbance spectra of first of two Elution 1 samples (containing pGEM-gbr22 protein) at 280 nm wavelength. Absorbance done at 10 mm pathlength. Through the use of Beer's Law (A=Ebc), with an extinction coefficient of 38,850, a pathlength of 1 cm, and an absorbance of 0.337, a concentration (c) of 8.67E-6 M was determined. Multiplying by the known molecular weight of the protein (25,794.2 g/mol), the concentration was also determined to be 0.224 mg/ml.
Figure 8: Image of Nanodrop spectrophotometer absorbance spectra of second of two Elution 1 samples (containing pGEM-gbr22 protein) at 280 nm wavelength. Absorbance reading done at 10 mm pathlength. Through the use of Beer's Law (A=Ebc), with an extinction coefficient of 38,850, a pathlength of 1 cm, and an absorbance of 0.389, a concentration of 1.001E-5 M was determined. Multiplying by the known molecular weight of the protein (25,794.2 g/mol), the concentration was also determined to be 0.258 mg/ml.
Figure 9: Image of Nanodrop spectrophotometer absorbance spectra of first of two Elution 1 samples at 574 nm maximum pGEM-gbr22 protein wavelength. Absorbance done at 1 mm pathlength. Through the use of Beer's Law (A=Ebc), with an extinction coefficient of 118,300, a pathlength of 0.1 cm, and an absorbance of 0.095, a concentration of 8.030E-6 M was determined. Multiplying by the known molecular weight of the protein (25,794.2 g/mol), the concentration was also determined to be 0.207 mg/ml.
Figure 10: Image of Nanodrop spectrophotometer absorbance spectra of second of two Elution 1 samples at 574 nm maximum pGEM-gbr22 protein wavelength. Absorbance done at 1 mm pathlength. Through the use of Beer's Law (A=Ebc), with an extinction coefficient of 118,300, a pathlength of 0.1 cm, and an absorbance of 0.089, a concentration of 7.523E-6 M was determined. Multiplying by the known molecular weight of the protein (25,794.2 g/mol), the concentration was also determined to be 0.194 mg/ml.
Figure 11: Failed attempt at Sodium Dodecyl Polyacrylamide Gel Electrophoresis (SDS-PAGE) of Fermentas Pageruler molecular weight standard, Samples 1-6, and a partner's Samples 4-6 due to running electrophoresis at 200 Volts for 25 minutes with gel cassette running in position which conducts minimal amount of current without another cassette in place as well.
Figure 12: Molecular weight standard for 4-20% TGS buffer when performing SDS-PAGE electrophoresis. Bands represent an estimate of the molecular weight (kDa) of proteins in a solution after SDS gel electrophoresis is done in this buffer.
Figure 13: Image of results of SDS gel electrophoresis ran for 25 minutes at 200 Volts after destaining, but prior to drying the gel. From left to right, the wells are represented by the Fermentas Pageruler molecular weights standard, Samples 1-6, and a partner's samples 4-6. Visible purple bands on the sample wells represent a distinguishable protein in the solution from which each of the samples came from. The intensity of the bands represent the prevalence of protein in solution, and their position in relation to the molecular weights standard can distinguish the relative molecular weight of the proteins.
Figure 14: Image of results of SDS gel electrophoresis ran for 25 minutes at 200 Volts after destaining and drying the gel. From left to right, the wells are represented by the Fermentas Pageruler molecular weights standard, Samples 1-6, and a partner's samples 4-6. Visible purple bands on the sample wells represent a distinguishable protein in the solution from which each of the samples came from. The intensity of the bands represent the prevalence of protein in solution, and their position in relation to the molecular weights standard can distinguish the relative molecular weight of the proteins. The most intense bands in the S5 and S6 sample solutions represent pGEM-gbr22 protein of interest.
Discussion:
After transforming the E. Coli BL21 (DE3) bacteria with the coral plasmid so that it produced the pGEM-gbr22 coral protein, the transformed bacteria was cultivated and grew into many colonies that expressed the purple protein. The bacteria that received the coral plasmid did not die off when incubated in a agar plate that contained ampicillin because the plasmid was genetically modified to code for ampicillin resistance in its host. The effects of the lack of an ampicillin resistant gene was seen when the E. Coli BL21 (DE3) bacteria that did not receive the plasmid did not grow into colonies. After further incubating the bacteria with LB and ampicillin in a water bath incubator for an extended period of time, Sample 1 was taken. Sample 1 consisted of E. Coli BL21 (DE3) cells with the recombinant plasmid included into its genome, and LB and ampicillin media. After growing up the bacterial cells to overexpress the pGEM-gbr22 protein, the cells were spun down through the use of an Allegra X-15 large benchtop centrifuge. After the supernatant was removed, lysozyme was added to the cellular solution to lyse the cell walls of the E. Coli bacteria so that the pGEM-gbr22 protein would then be a separate entity from the bacterial cell itself in solution (although still mixed with other proteins and many cellular debris still). Benzonase was then added to the solution in order to break up the DNA/RNA in the solution so that the solution would have reduced viscosity for easier protein purification. The solution then underwent centrifugation again so that the insoluble cellular debris that were not of any use were part of the pellet. Sample 2 was taken after this step, and it consisted of the supernatant from the centrifugation that contained mainly the soluble proteins of what was once inside the E. Coli bacterial cells. The soluble proteins were then syringe filtered to rid of the larger macromolecular debris, and then Ni-NTA resin/buffer was added to the resulting solution. This resin had a specific affinity to proteins with multi-histidine residues in a row together. Since the pGEM-gbr22 plasmid was genetically modified to contain a hexa-histidine continuous chain when the protein was produced, the hexa-histidine tag had an affinity to the Ni-NTA resin beads. When the Ni-NTA column was set up and the soluble protein solution with the Ni-NTA resin was poured into the column, the soluble proteins that did not have a binding affinity towards the Ni-NTA resin slowly dripped out of the column into a 'waste' tube. Sample 3 was taken from this waste tube, and it contained soluble proteins that the E. Coli synthesized that did not have a long continuous chain of histidine residues. A Wash buffer (which contained small traces of imidazole) was then made to wash off some of the loosely bound proteins that may have bound to the Ni-NTA resin by random chance; this was what Sample 4 consisted of. A Elution buffer (high amount of imidazole) was then made and sent through the column. The imidazole binded to the Ni-NTA resin, which replaced the binding of the hexa-histidine tag in the pGEM-gbr22 protein, thus allowing the protein to fall out of the column and be present in a tube labeled 'Elution 1'; this was what Sample 5 consisted of (the majority of the pGEM-gbr22 protein). Sample 6 consisted of a second pass of the Elution buffer so that any pGEM-gbr22 proteins that did not detach from the Ni-NTA resin from the first Elution pass would most likely detach on the second pass. Through the use of Nanodrop spectrophotogaphy, the Elution 1 tube containing the pGEM-gbr22 protein was further analyzed for the absorbance of the protein at the arbitrary 280 nm wavelength and its maximal wavelength (574 nm) [4]. Through the absorbance readings and the knowledge of the extinction coefficients of the protein at 280 nm (from http://ca.expasy.org/tools/protparam.html) and at 574 nm [4], Beer's Law was applied to determine the concentration of the pGEM-gbr22 protein at these different wavelengths (A=Ebc). The average concentration found for a wavelength of 280 nm was 0.2409 mg/ml, and the average concentration found for a maximal wavelength of 574 nm was 0.2005 mg/ml. Since 6.0 ml of purified pGEM-gbr22 protein was collected, a yield of 1.445 mg was found using 280 nm wavelength, and a yield of 1.2054 mg was found using the maximal 574 nm wavelength. The six samples were then characterized through the use of Sodium Dodecyl Polycrylamide Gel Elextrophoresis (SDS-PAGE). The most distinctly visible bands in Samples 5 and 6 should have represented the pGEM-gbr22 purple coral protein which was most prevalent in solution, since it was determined that Sample 5 contained the majority of the pGEM-gbr22 purple coral protein, and Sample 6 contained the mainly remnants of the protein. It was expected that the most distinctly intense band in Sample 5 also was to be the most distinctly intense band in Sample 6, and this held true based on Figure 14. A few other slightly intense bands still showed up in Samples 5 and 6. No matter how hard one may try, a single protein cannot be singly purified and isolated because two proteins may have enough similar characteristics that allow them to pass through each step of the purification process in certain amounts. Since other bands do exist in the gel for Samples 5 and 6, a ballpark estimate of about 65% pGEM-gbr22 protein can be determined for these samples. By comparing the bands to the molecular weight standard in the leftmost well of figure 14, and then comparing that to figure 12, it was determined that the molecular weight of the pGEM-gbr22 protein was about 29 kDa. Within the process of expressing, purifying, and characterizing the pGEM-gbr22 protein of interest, many sources of error may have occurred along the way. Small errors, such as in pipetting and mixing, may have occurred numerous amounts of times, but there were two main sources of error that were experienced altogether. First, when analyzing the absorbance of the pGEM-gbr22 protein at maximal wavelength through the use of Nanodrop, the absorbance data in figures 9 and 10 were not smooth and instead were extremely jagged.This may have been due to faulty readings by the spectrophotometer itself, or improper blanking or cleaning techniques were used while running the spectrophotometer. A second major source of error included running the gel wrong in the electrophoresis module. Without two gel cassettes in the electrophoresis tank, it must be ensured that a single cassette is placed where the tank conducts the most current. If the gel cassette is placed in the wrong position without another cassette present, the gel will not have enough voltage to be run successfully.
Conclusions:
Through overexpression of a purple coral pGEM-gbr22 protein in E. Coli BL21 (DE3) bacteria, the protein was then isolated and purified in a solution. The concentration of the purified protein in solution could be determined through the use of Beer's Law, and the molecular weight and purity of the protein in relation to other proteins in solution could be determined through the use of SDS gel electrophoresis. It was found that a protein may be purified as much as possible, but there will still typically be a small amount of a few other proteins in solution as well. In the future of the Virtual Drug Screening stream, certain proteins from a bacteria, virus, or fungi may be purified so that a potential drug may target the protein, and then it can be determined easily whether the drug worked in inhibiting the protein or not.
References:
[1] Gräslund, S.; Nordlund, P.; Weigelt, J.; Hallberg, B. M.; Bray, J.; Gileadi, O.; Knapp, S.; Oppermann, U.; Arrowsmith, C.; Hui, R.; Ming, J.; dhe-Paganon, S.; Park, H. W.; Savchenko, A.; Yee, A.; Edwards, A.; Vincentelli, R.; Cambillau, C.; Kim, R.; Kim, S. H.; Rao, Z.; Shi, Y.; Terwilliger, T. C.; Kim, C. Y.; Hung, L. W.; Waldo, G. S.; Peleg, Y.; Albeck, S.; Unger, T.; Dym, O.; Prilusky, J.; Sussman, J. L.; Stevens, R. C.; Lesley, S. A.; Wilson, I. A.; Joachimiak, A.; Collart, F.; Dementieva, I.; Donnelly, M. I.; Eschenfeldt, W. H.; Kim, Y.; Stols, L.; Wu, R.; Zhou, M.; Burley, S. K.; Emtage, J. S.; Sauder, J. M.; Thompson, D.; Bain, K.; Luz, J.; Gheyi, T.; Zhang, F.; Atwell, S.; Almo, S. C.; Bonanno, J. B.; Fiser, A.; Swaminathan, S.; Studier, F. W.; Chance, M. R.; Sali, A.; Acton, T. B.; Xiao, R.; Zhao, L.; Ma, L. C.; Hunt, J. F.; Tong, L.; Cunningham, K.; Inouye, M.; Anderson, S.; Janjua, H.; Shastry, R.; Ho, C. K.; Wang, D.; Wang, H.; Jiang, M.; Montelione, G. T.; Stuart, D. I.; Owens, R. J.; Daenke, S.; Schütz, A.; Heinemann, U.; Yokoyama, S.; Büssow, K.; Gunsalus, K. C.; Consortium, S. G.; Consortium, C. S. G.; Consortium, N. S. G., Protein production and purification. Nat Methods2008,5 (2), 135-46.
[2] Guo, F.; Zhu, G., Presence and removal of a contaminating NADH oxidation activity in recombinant maltose-binding protein fusion proteins expressed in Escherichia coli.Biotechniques2012,52 (4), 247-53.
[3] Nilvebrant, J.; Alm, T.; Hober, S., Orthogonal protein purification facilitated by a small bispecific affinity tag. J Vis Exp2012, (59).
[4] Alieva, N. O.; Konzen, K. A.; Field, S. F.; Meleshkevitch, E. A.; Hunt, M. E.; Beltran-Ramirez, V.; Miller, D. J.; Wiedenmann, J.; Salih, A.; Matz, M. V., Diversity and evolution of coral fluorescent proteins. PLoS ONE 2008, 3, (7), e2680.
The Expression, Purification, and Characterization of pGEM-gbr22 Recombinant Protein
Introduction:
Many different types of organisms have been used to synthesize proteins from another organism of interest. The relatively harmless strain of competent bacterial E. Coli BL21 (DE3) has consistently proven itself as one of the optimal hosts for expression of proteins of higher level organisms due to a rapid rate of bacterial replication and a lack of Ion and ompT proteases and its compatibility with the T7 lacO promoter system [1]. Recombinant protein plasmids have often been tagged with a hexa-histidine affinity tag for future ease of purification of the protein, and a ampicillin resistant gene has often been inserted into the plasmid as well [2]. The recombinant plasmid may then be inserted into the E. Coli BL21 (DE3) bacterial genome, and the bacteria can be cultivated to express the protein from the plasmid. After protein expression, the bacterial cells could then become lysed, and the protein of interest could run through a column containing Ni-NTA resin so that the hexa-histidine affinity tag would bind to the resin and the protein of interest would then be successfully isolated and purified [3]. For a basic introduction into protein expression and purification, a visibly purple and modified (ampicillin resistant gene and insertion of hexa-histidine tag) pGEM-gbr22 Great Barrier Reef coral protein was expressed in the E. Coli BL21 (DE3) bacteria. After cultivation and successful protein production, the pGEM-gbr22 protein of interest was purified through the use of a Ni-NTA resin column, and then the protein was later characterized through the use of gel electrophoresis. It can be hypothesized that the the pGEM-bgr22 protein will ultimately be isolated from its host E. Coli bacteria, but it can never be fully isolated from some of the contents of the host bacteria in which it grew in.
Materials & Methods:
While in proximity of a lit gas Bunsen burner, 25 microliters of competent E. Coli BL21 (DE3) bacterial cells were added into one transformation tube, and the same amount of bacterial cells were added into another transformation tube. 1.29 microliters of pGEM-gbr22 plasmid was then carefully pipetted into only one of the transformation tube. After the two transformation tubes were placed on ice for 30 minutes, the two tubes were heat shocked for exactly 45 seconds in a 42 degree Celsius water bath. The two tubes were then placed in ice for another 2 minutes. 200 microliters of SOC media was then pipetted into the two tubes, then the tubes were placed in a water bath incubator (~250 RPM) for 30 minutes at 37 degrees Celsius. Next, 5 colirollers were placed onto a agar ampicillin plate labeled 'DNA', and 5 colirollers were placed onto a seperate agar ampicillin plate labeled 'No DNA (Control)'. After the transformation tubes finished their 30 minutes in the water bath incubator, the 50 microliters of the bacteria/SOC mixture was pipetted from the tube with the plasmid into the 'DNA' plate. 50 microliters of the bacteria/SOC mixture was then pipetted from the tube without the plasmid into the 'No DNA (Control)' plate. The mixtures were then rolled around in their respective plates, the colirollers were then carefully poured out, and then the plates were placed in a 37 degree Celsius incubator overnight. A fun plate with microorganisms from the men's restroom was also created and incubated overnight. On the next morning, the plates were taken from the incubator, and 0.01 ml of ampicillin stock (50 mg/ml) was added to two 14 ml round bottom conical tubes with 5 ml of LB in it. Two sterile small pipette small were then used to gently swipe and pick up each a single colony of transformed bacteria from the 'DNA' plate. One tip (with the bacterial colony on it) was then entirely submerged in one of the tubes of LB and ampicillin media, and the other tip was then entirely submerged in the second tube of LB and ampicillin media as a backup. These two conical tubes were then placed in a shaking water bath incubator (~250-300 RPM) at 37 degrees Celsius for 8 hours. After 8 hours, the two tubes were removed from the water bath, and one 125 ml Erlenmeyer Flask was filled with 25 ml of LB media and of 0.05 ml ampicillin. Exactly 0.625 ml of the culture from one of the tubes (the most visibly purple tube) that has just been removed from the water bath was then pipetted into the Erlenmeyer Flask. Foil was placed over the top of the flask, and the flask was placed back into the shaking water bath incubator at the same settings as before, but 16-24 hours of growth was then allowed. After the allotted 16-24 hours of incubation, the now extremely visibly purple flask was removed from the incubator, and 500 ml of the mixture in the flask was pipetted into a 1.7 ml Eppendorf tube and saved in 4 degree Celsius refrigerator labeled as Sample 1. The bacteria and the media from the flask were then poured in its entirety into a 50 ml conical tube. After securely capping the tube, it was then placed into the Allegra X-15 (Beckman Coulter, Inc., Brea, CA) benchtop centrifuge with another equally weighted conical tube of the same mass in opposite position for balancing purposes. The tube was centrifuged at 5,000 RPM for 10 minutes at 4 degrees Celsius. After centrifuging, the liquid supernatant was removed from the tube and was decanted into a bleach waste container. The remaining purple cell pellet was pipetted up and down with 2.5 ml of 1x PBS solution to form a homogenous suspension. 50 microliters of lysozyme was then added to the conical tube was well, and once vortexed, it was placed in a -20 degree Celcius freezer. After a week or so, the 50 ml conical tube containing the lysozyme and the 2.5 ml of suspended bacterial cells was thawed, and then 2 microliters of Benzonase (Sigma-Aldrich, St. Louis, MO) was added to the conical tube. It was mixed by inversion for about 15 minutes at room temperature. After this, a 1 ml of the mixture was pipetted in the conical tube into two 1.7 ml Eppendorf tubes, and the pipette the rest into a third 1.7 ml Eppendorf tube. The three Eppendorf tubes were then centrifuged in the small centrifuge for 20 minutes at 14,000 RPM at 4 degrees Celsius. After centrifuging, 50 microliters of the supernatant from one of the tubes was stored and labeled as Sample 2. The bacterial cell debris was the resulting pellet from centrifuging, so the liquid supernatant (containing soluable proteins) from each of the three tubes was extracted through pipetting, and then it was placed in a clean 15 ml conical tube. A 10 ml Wash buffer containing 1.0 ml of 1x PBS and 0.2 ml of imidazole, and a 10 ml Elution buffer containing 1.0 ml of 1x PBS and 2.5 ml of imidazole were then created and kept on ice. The supernatant was then filtered through a PES syringe filter (Membrane Solutions, Plano, TX). A purification column was then set up by first adding 0.5 ml of Ni-NTA resin/buffer mixture into the conical tube of supernatant. Through occasional inversion mixing, this new mixture was incubated at room temperature for 15 minutes. A 20 ml purification chromatography Econo column (Bio-Rad, Hercules, CA) was then set up on a ring stand with a ice bucket below it. The column was first rinsed with nanopure water, then the supernatant with the Ni-NTA resin/buffer was ran through the column, allowed to settle for about 5 minutes, and then it was dripped into a 10 ml round bottom conical tube labeled 'waste'. A 50 microliter sample of waste solution was taken and labeled Sample 3 and stored in a 4 degree Celsius refrigerator. It was crucial to leave a small amount of the 'waste' solution in the column so that not all of it flowed out of the column (the protein of interest settled to the bottom of the column). After re-plugging the column, 5 ml of the Wash solution was ran through the column into a new 10 ml conical tube labeled 'Wash' after about 5 minutes of letting the Wash solution settle. A 50 microliter sample of Wash solution was taken and labeled as Sample 4 and stored in a 4 degree Celsius refrigerator. After leaving a small amount of solution in the column, 5 ml of the Elution solution was ran through the column in the same way as before into a 'Elution 1' 15 ml conical tube. A 50 microliter Sample 5 was taken from this Elution 1 tube and stored in a 4 degree Celsius refrigerator. Another 5 ml of Elution solution was ran through the column into a 'Elution 2' 15 ml conical tube. A 50 microliter Sample 6 was taken from this Elution 2 tube and stored in a 4 degree Celsius refrigerator. To strip the Ni-NTA from the column, 10 column volumes (cv) of water, then 10 cv's of 0.5 NaOH, then 10 cv's of water were ran through the column one at a time, and then the column was stored in a 4 degree Celsius refrigerator with 1 ml of 30% ethanol in nanopure water sitting on the bottom of the column. In order to estimate the concentration of the pGEM-gbr22 purple protein, a Nanodrop (Thermo Scientific, Wilmington, DE) spectrophotometer was used to measure the absorbance of the protein at 280 nm and the maximal wavelength of the protein (574 nm). This was done by first blanking the spectrophotomer with 2 microliters of nanopure water, then blanking the instrument with 2 microliters of Elution solution buffer, and then measuring the protein's absorbance by adding 2 microliters of Elution 1 to the spectrophotometer (ensuring a Kimwipe was used to scrub the pedestal after each step). In the process of characterizing the protein, 10 microliters of a 6x gel loading buffer was added to the saved Samples 2-6 and pipetted up and down to mix well. Sample 1 was centrifuged at 5,000 RPM for 5 minutes using a small centrifuge with a 500 microliter Eppendorf tube of water as a counterbalance. The supernatant was removed after centrifuging, and then 200 microliters of water and 40 microliters of the 6x gel loading buffer were added to the cell pellet and mixed to create a homogeneous solution. Then all the samples (1-6) were placed on a 95 degree Celsius heat block for 5 minutes and then centrifuged at 5,000 RPM for 2 minutes. An SDS electrophoresis module was then assembled and a 1x TGS buffer was used to fill the tank. After the 10 wells were cleaned using a small needle and about 2 ml of the 1x TGS buffer and the tape on the bottom of the gel was removed, 7 microliters of a Fermentas prestained protein ladder (Thermo Fisher Scientific, Waltham, MA) was pipetted into the first well, and then wells 2-7 were filled with 20 microliters of pipetted Samples 1-6. Wells 8-10 were filled with a partner's Samples 4-6. The electrophoresis module was then ran for 25 minutes at 200 Volts. When finished, the gel was removed from its casing and allowed to float in a small plastic dish of nanopure water. The SDS residue on the gel was then removed by washing the gel three times in a orbital shaker with about 100 ml of nanopure water for five minutes each time. After the water was removed from the plastic container, an Imperial protein stain (Pierce Biotechnology, Rockford, IL) was then added to completely cover the gel, and then the container was placed back on the orbital shaker for about an hour. After an hour, the stain was poured back into the container from which it originated, and the gel was washed twice (as described previously), and a folded kimwipe was added to the container filled with clean water and gel for overnight orbital shaking. On the next day, the gel was removed from the plastic container, placed ontop a rectangle of Whatman filter paper (GE Healthcare, Maidstone, UK), and a piece of rectangular cellophane was then placed on top of the gel. The gel was then placed on a drying bed for 1.5 hours at 75 degrees Celsius on the Gradient cycle.
Results:
Figure 1: Image of "Fun Plate" bacterial, virus, and fungal colonies. Microorganisms collected from men's restroom toilet handle. Incubated in 37 degree Celsius incubator for about 24 hours. Ampicillin not present on agar of plate.
Figure 2: Image of E. Coli BL21 (DE3) bacterial colonies incubated for about 24 hours in 37 degree Celsius incubator in the presence of ampicillin added to the agar. Very few colonies developed due to the ampicillin's antibiotic effect that killed off most of the bacterial colonies during incubation.
Figure 3: Image of E. Coli BL21 (DE3) bacterial colonies incubated for about 24 hours in 37 degree Celsius incubator. These bacterium have been manipulated to express an ampicillin resistant purple Great Barrier Reef coral gene (pGEM-gbr22) in the form of a protein. There are numerous bacterial colonies despite there being ampicillin in the agar because the bacteria with the included coral gene now express an ampicillin resistant gene. Most other bacterial colonies that may have developed in the plate died during incubation due to the presence of ampicillin in the agar.
Figure 4: Image of E. Coli BL21 (DE3) bacteria with expressed pGEM-gbr22 protein resting in an Erlenmeyer Flask mixed with 25 ml LB broth and ampicillin media after about 24 hours of incubation in 37 degree Celsius shaking water bath incubator.
Figure 5: Image of spun down (from Allegra X-15 benchtop centrifuge) E. Coli BL21 (DE3) bacteria with expressed pGEM-gbr22 protein pellet with the supernatant removed from the 50 ml conical tube. Wet cell pellet weighed about 0.24 g.
Figure 6: Image of Elution 1 (bulk of purified pGEM-gbr22 protein washed down by Elution buffer solution in Ni-NTA column) and Elution 2 (remnants of purified pGEM-gbr22 protein that did not drip into Elution 1 tube from the Ni-NTA column) of purified pGEM-gbr22 protein in two 15 mL conical tubes.
Figure 7: Image of Nanodrop spectrophotometer absorbance spectra of first of two Elution 1 samples (containing pGEM-gbr22 protein) at 280 nm wavelength. Absorbance done at 10 mm pathlength. Through the use of Beer's Law (A=Ebc), with an extinction coefficient of 38,850, a pathlength of 1 cm, and an absorbance of 0.337, a concentration (c) of 8.67E-6 M was determined. Multiplying by the known molecular weight of the protein (25,794.2 g/mol), the concentration was also determined to be 0.224 mg/ml.
Figure 8: Image of Nanodrop spectrophotometer absorbance spectra of second of two Elution 1 samples (containing pGEM-gbr22 protein) at 280 nm wavelength. Absorbance reading done at 10 mm pathlength. Through the use of Beer's Law (A=Ebc), with an extinction coefficient of 38,850, a pathlength of 1 cm, and an absorbance of 0.389, a concentration of 1.001E-5 M was determined. Multiplying by the known molecular weight of the protein (25,794.2 g/mol), the concentration was also determined to be 0.258 mg/ml.
Figure 9: Image of Nanodrop spectrophotometer absorbance spectra of first of two Elution 1 samples at 574 nm maximum pGEM-gbr22 protein wavelength. Absorbance done at 1 mm pathlength. Through the use of Beer's Law (A=Ebc), with an extinction coefficient of 118,300, a pathlength of 0.1 cm, and an absorbance of 0.095, a concentration of 8.030E-6 M was determined. Multiplying by the known molecular weight of the protein (25,794.2 g/mol), the concentration was also determined to be 0.207 mg/ml.
Figure 10: Image of Nanodrop spectrophotometer absorbance spectra of second of two Elution 1 samples at 574 nm maximum pGEM-gbr22 protein wavelength. Absorbance done at 1 mm pathlength. Through the use of Beer's Law (A=Ebc), with an extinction coefficient of 118,300, a pathlength of 0.1 cm, and an absorbance of 0.089, a concentration of 7.523E-6 M was determined. Multiplying by the known molecular weight of the protein (25,794.2 g/mol), the concentration was also determined to be 0.194 mg/ml.
Figure 11: Failed attempt at Sodium Dodecyl Polyacrylamide Gel Electrophoresis (SDS-PAGE) of Fermentas Pageruler molecular weight standard, Samples 1-6, and a partner's Samples 4-6 due to running electrophoresis at 200 Volts for 25 minutes with gel cassette running in position which conducts minimal amount of current without another cassette in place as well.
Figure 12: Molecular weight standard for 4-20% TGS buffer when performing SDS-PAGE electrophoresis. Bands represent an estimate of the molecular weight (kDa) of proteins in a solution after SDS gel electrophoresis is done in this buffer.
Figure 13: Image of results of SDS gel electrophoresis ran for 25 minutes at 200 Volts after destaining, but prior to drying the gel. From left to right, the wells are represented by the Fermentas Pageruler molecular weights standard, Samples 1-6, and a partner's samples 4-6. Visible purple bands on the sample wells represent a distinguishable protein in the solution from which each of the samples came from. The intensity of the bands represent the prevalence of protein in solution, and their position in relation to the molecular weights standard can distinguish the relative molecular weight of the proteins.
Figure 14: Image of results of SDS gel electrophoresis ran for 25 minutes at 200 Volts after destaining and drying the gel. From left to right, the wells are represented by the Fermentas Pageruler molecular weights standard, Samples 1-6, and a partner's samples 4-6. Visible purple bands on the sample wells represent a distinguishable protein in the solution from which each of the samples came from. The intensity of the bands represent the prevalence of protein in solution, and their position in relation to the molecular weights standard can distinguish the relative molecular weight of the proteins. The most intense bands in the S5 and S6 sample solutions represent pGEM-gbr22 protein of interest.
Discussion:
After transforming the E. Coli BL21 (DE3) bacteria with the coral plasmid so that it produced the pGEM-gbr22 coral protein, the transformed bacteria was cultivated and grew into many colonies that expressed the purple protein. The bacteria that received the coral plasmid did not die off when incubated in a agar plate that contained ampicillin because the plasmid was genetically modified to code for ampicillin resistance in its host. The effects of the lack of an ampicillin resistant gene was seen when the E. Coli BL21 (DE3) bacteria that did not receive the plasmid did not grow into colonies. After further incubating the bacteria with LB and ampicillin in a water bath incubator for an extended period of time, Sample 1 was taken. Sample 1 consisted of E. Coli BL21 (DE3) cells with the recombinant plasmid included into its genome, and LB and ampicillin media. After growing up the bacterial cells to overexpress the pGEM-gbr22 protein, the cells were spun down through the use of an Allegra X-15 large benchtop centrifuge. After the supernatant was removed, lysozyme was added to the cellular solution to lyse the cell walls of the E. Coli bacteria so that the pGEM-gbr22 protein would then be a separate entity from the bacterial cell itself in solution (although still mixed with other proteins and many cellular debris still). Benzonase was then added to the solution in order to break up the DNA/RNA in the solution so that the solution would have reduced viscosity for easier protein purification. The solution then underwent centrifugation again so that the insoluble cellular debris that were not of any use were part of the pellet. Sample 2 was taken after this step, and it consisted of the supernatant from the centrifugation that contained mainly the soluble proteins of what was once inside the E. Coli bacterial cells. The soluble proteins were then syringe filtered to rid of the larger macromolecular debris, and then Ni-NTA resin/buffer was added to the resulting solution. This resin had a specific affinity to proteins with multi-histidine residues in a row together. Since the pGEM-gbr22 plasmid was genetically modified to contain a hexa-histidine continuous chain when the protein was produced, the hexa-histidine tag had an affinity to the Ni-NTA resin beads. When the Ni-NTA column was set up and the soluble protein solution with the Ni-NTA resin was poured into the column, the soluble proteins that did not have a binding affinity towards the Ni-NTA resin slowly dripped out of the column into a 'waste' tube. Sample 3 was taken from this waste tube, and it contained soluble proteins that the E. Coli synthesized that did not have a long continuous chain of histidine residues. A Wash buffer (which contained small traces of imidazole) was then made to wash off some of the loosely bound proteins that may have bound to the Ni-NTA resin by random chance; this was what Sample 4 consisted of. A Elution buffer (high amount of imidazole) was then made and sent through the column. The imidazole binded to the Ni-NTA resin, which replaced the binding of the hexa-histidine tag in the pGEM-gbr22 protein, thus allowing the protein to fall out of the column and be present in a tube labeled 'Elution 1'; this was what Sample 5 consisted of (the majority of the pGEM-gbr22 protein). Sample 6 consisted of a second pass of the Elution buffer so that any pGEM-gbr22 proteins that did not detach from the Ni-NTA resin from the first Elution pass would most likely detach on the second pass. Through the use of Nanodrop spectrophotogaphy, the Elution 1 tube containing the pGEM-gbr22 protein was further analyzed for the absorbance of the protein at the arbitrary 280 nm wavelength and its maximal wavelength (574 nm) [4]. Through the absorbance readings and the knowledge of the extinction coefficients of the protein at 280 nm (from http://ca.expasy.org/tools/protparam.html) and at 574 nm [4], Beer's Law was applied to determine the concentration of the pGEM-gbr22 protein at these different wavelengths (A=Ebc). The average concentration found for a wavelength of 280 nm was 0.2409 mg/ml, and the average concentration found for a maximal wavelength of 574 nm was 0.2005 mg/ml. Since 6.0 ml of purified pGEM-gbr22 protein was collected, a yield of 1.445 mg was found using 280 nm wavelength, and a yield of 1.2054 mg was found using the maximal 574 nm wavelength. The six samples were then characterized through the use of Sodium Dodecyl Polycrylamide Gel Elextrophoresis (SDS-PAGE). The most distinctly visible bands in Samples 5 and 6 should have represented the pGEM-gbr22 purple coral protein which was most prevalent in solution, since it was determined that Sample 5 contained the majority of the pGEM-gbr22 purple coral protein, and Sample 6 contained the mainly remnants of the protein. It was expected that the most distinctly intense band in Sample 5 also was to be the most distinctly intense band in Sample 6, and this held true based on Figure 14. A few other slightly intense bands still showed up in Samples 5 and 6. No matter how hard one may try, a single protein cannot be singly purified and isolated because two proteins may have enough similar characteristics that allow them to pass through each step of the purification process in certain amounts. Since other bands do exist in the gel for Samples 5 and 6, a ballpark estimate of about 65% pGEM-gbr22 protein can be determined for these samples. By comparing the bands to the molecular weight standard in the leftmost well of figure 14, and then comparing that to figure 12, it was determined that the molecular weight of the pGEM-gbr22 protein was about 29 kDa. Within the process of expressing, purifying, and characterizing the pGEM-gbr22 protein of interest, many sources of error may have occurred along the way. Small errors, such as in pipetting and mixing, may have occurred numerous amounts of times, but there were two main sources of error that were experienced altogether. First, when analyzing the absorbance of the pGEM-gbr22 protein at maximal wavelength through the use of Nanodrop, the absorbance data in figures 9 and 10 were not smooth and instead were extremely jagged.This may have been due to faulty readings by the spectrophotometer itself, or improper blanking or cleaning techniques were used while running the spectrophotometer. A second major source of error included running the gel wrong in the electrophoresis module. Without two gel cassettes in the electrophoresis tank, it must be ensured that a single cassette is placed where the tank conducts the most current. If the gel cassette is placed in the wrong position without another cassette present, the gel will not have enough voltage to be run successfully.
Conclusions:
Through overexpression of a purple coral pGEM-gbr22 protein in E. Coli BL21 (DE3) bacteria, the protein was then isolated and purified in a solution. The concentration of the purified protein in solution could be determined through the use of Beer's Law, and the molecular weight and purity of the protein in relation to other proteins in solution could be determined through the use of SDS gel electrophoresis. It was found that a protein may be purified as much as possible, but there will still typically be a small amount of a few other proteins in solution as well. In the future of the Virtual Drug Screening stream, certain proteins from a bacteria, virus, or fungi may be purified so that a potential drug may target the protein, and then it can be determined easily whether the drug worked in inhibiting the protein or not.
References:
[1] Gräslund, S.; Nordlund, P.; Weigelt, J.; Hallberg, B. M.; Bray, J.; Gileadi, O.; Knapp, S.; Oppermann, U.; Arrowsmith, C.; Hui, R.; Ming, J.; dhe-Paganon, S.; Park, H. W.; Savchenko, A.; Yee, A.; Edwards, A.; Vincentelli, R.; Cambillau, C.; Kim, R.; Kim, S. H.; Rao, Z.; Shi, Y.; Terwilliger, T. C.; Kim, C. Y.; Hung, L. W.; Waldo, G. S.; Peleg, Y.; Albeck, S.; Unger, T.; Dym, O.; Prilusky, J.; Sussman, J. L.; Stevens, R. C.; Lesley, S. A.; Wilson, I. A.; Joachimiak, A.; Collart, F.; Dementieva, I.; Donnelly, M. I.; Eschenfeldt, W. H.; Kim, Y.; Stols, L.; Wu, R.; Zhou, M.; Burley, S. K.; Emtage, J. S.; Sauder, J. M.; Thompson, D.; Bain, K.; Luz, J.; Gheyi, T.; Zhang, F.; Atwell, S.; Almo, S. C.; Bonanno, J. B.; Fiser, A.; Swaminathan, S.; Studier, F. W.; Chance, M. R.; Sali, A.; Acton, T. B.; Xiao, R.; Zhao, L.; Ma, L. C.; Hunt, J. F.; Tong, L.; Cunningham, K.; Inouye, M.; Anderson, S.; Janjua, H.; Shastry, R.; Ho, C. K.; Wang, D.; Wang, H.; Jiang, M.; Montelione, G. T.; Stuart, D. I.; Owens, R. J.; Daenke, S.; Schütz, A.; Heinemann, U.; Yokoyama, S.; Büssow, K.; Gunsalus, K. C.; Consortium, S. G.; Consortium, C. S. G.; Consortium, N. S. G., Protein production and purification. Nat Methods 2008, 5 (2), 135-46.
[2] Guo, F.; Zhu, G., Presence and removal of a contaminating NADH oxidation activity in recombinant maltose-binding protein fusion proteins expressed in Escherichia coli.Biotechniques 2012, 52 (4), 247-53.
[3] Nilvebrant, J.; Alm, T.; Hober, S., Orthogonal protein purification facilitated by a small bispecific affinity tag. J Vis Exp 2012, (59).
[4] Alieva, N. O.; Konzen, K. A.; Field, S. F.; Meleshkevitch, E. A.; Hunt, M. E.; Beltran-Ramirez, V.; Miller, D. J.; Wiedenmann, J.; Salih, A.; Matz, M. V., Diversity and evolution of coral fluorescent proteins. PLoS ONE 2008, 3, (7), e2680.